LM317 experiments and measurements - Page 3 - diyAudio
Go Back   Home > Forums > Amplifiers > Power Supplies

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 18th May 2011, 10:15 PM   #21
just another
diyAudio Moderator
 
wintermute's Avatar
 
Join Date: Aug 2003
Location: Sydney
Blog Entries: 22
Thanks George, I was completely unaware of the effects on the bandwidth. I could add an additional toggle switch to switch to switch between the 100K and 1M resistors. I'm Mostly interested in the audio spectrum so 34Khz should be fine, and I can swap back to 10 X gain to look at the higher frequencies (up to about 86Khz with 192Khz sampling rate). My audigy only does 96Khz sampling so that tops out at about 48Khz., and I will probably be able to use the audigy with 100 X gain.

Tony.
__________________
Any intelligence I may appear to have is purely artificial!
Some of my photos
  Reply With Quote
Old 19th May 2011, 06:48 AM   #22
gpapag is offline gpapag  Greece
diyAudio Member
 
gpapag's Avatar
 
Join Date: Nov 2002
Location: Athens-Greece
Hi Tony

I have some bad news and some good news for you.

I start with the bad news.
What I wrote you above are valid for TL074 but not for LM324 (which I don’t have).

But from the datasheets, one notes that:
LM324 has a unity gain bandwidth of 1MHz.
TL074 has a unity gain bandwidth of 3MHz.

This means that for the “preferred” configuration for X100 gain
C3=10uF
R5=1M
R6=10k

the LM324 will have a HF –3dB of 11.6kHz, that is 1/3 of TL074.
This is too limited a HF response.

Now the good news.

Replace your LM324 with the NE5514.
It is pin compatible and recommended in the datasheet as a replacement to LM324.
It is cheap and easy to find.
NE5514 has a unity gain bandwidth of 3MHz.

This way, you are back to the HF –3dB of 35kHz for a gain of X100 (40dB).

Regards
George

*PS
You may find other quad op amps that go higher than NE5514 (i.e. op amps that have a higher unity gain bandwidth).
But these require a very careful lay-out and power supply local decoupling, or else they will oscillate.
With the NE5514 I hope that you will not experience such oscillation problems with your existing pcb.
__________________
["Second Law is a bitch." - SY] ["The Road To Heaven:Specify the performance & accept the design. The Road To Hell:Specify the design & accept the performance"-Bruno Putzeys]

Last edited by gpapag; 19th May 2011 at 07:10 AM.
  Reply With Quote
Old 19th May 2011, 09:19 AM   #23
diyAudio Member
 
digits's Avatar
 
Join Date: Apr 2011
WOuld you do a test with say two 1000u caps in parallel, but seperated by a low ohm resistor bypassed by a small inductor (few turns of wire around the resistor)....after the reg, vs, just the two in parallel.
  Reply With Quote
Old 19th May 2011, 11:30 AM   #24
just another
diyAudio Moderator
 
wintermute's Avatar
 
Join Date: Aug 2003
Location: Sydney
Blog Entries: 22
Thanks for all this George! I'd almost certainly have been disappointed if I'd taken my naive approach of simply changing the feedback resistors! in a bizarre twist my original gut feeling that the circuit might not be up to it was correct

I had a look for the NE5514 but its not at the local suppliers, TL074 is however pin compatible (even if it is jfet input compared to bipolar) will happily run at my voltage of 12V, and is abundantly available (though at rather varying prices, from about 56c to $3.50!) of course the convenient ones are $3.50 Only problem is I'm not sure if it is unity gain stable for the non 100X modes of the preamp...
Could have to put the preamp upgrade on hold for a while, and in some respects I might be better off designing something new.

I was planning on doing some more measurements tonight but its getting a bit late so will probably won't be as many as I was intending to.

The attachment is the board layout, I don't know enough about layout to know whether it will be good/bad/indifferent for a higher bandwidth opamp.

Tony.
Attached Images
File Type: jpg DSC_7459.jpg (309.9 KB, 704 views)
__________________
Any intelligence I may appear to have is purely artificial!
Some of my photos
  Reply With Quote
Old 19th May 2011, 01:14 PM   #25
gpapag is offline gpapag  Greece
diyAudio Member
 
gpapag's Avatar
 
Join Date: Nov 2002
Location: Athens-Greece
Quote:
I had a look for the NE5514 but its not at the local suppliers
I am surprised. Have a look if they have it as SE5514. It is the same chip.

Quote:
TL074 is however pin compatible (even if it is jfet input compared to bipolar) will happily run at my voltage of 12V, and is abundantly available (though at rather varying prices, from about 56c to $3.50!) of course the convenient ones are $3.50
Only problem is I'm not sure if it is unity gain stable for the non 100X modes of the preamp...
TL074 is unity gain stable. I have used it as such. Have a look at the data sheet, page 7 (and on).

http://www.datasheetcatalog.org/data...ents/tl074.pdf

What I don’t know is if it can be used with single power supply in your circuit without modifications to the pcb.
If you already bought one (I hope not for $3.50, this is an arm robbery) drop it in the circuit and see how it goes.
Apply a signal at circuit’s input, load the output with ~100k (do not have your soundcard connected to it yet) and check with your oscilloscope first for presence of dc at the output. Then check for obvious asymmetric clipping when you increase the input signal. (*PS)
If none of them is present, then switch off, remove the 100k load and connect the circuit to your sound card. Pretend that nothing has changed.

Quote:
The attachment is the board layout, I don't know enough about layout to know whether it will be good/bad/indifferent for a higher bandwidth opamp.
It is not OK for a high bandwidth opamps (but it will work with NE5514 and the likes)

Quote:
and in some respects I might be better off designing something new.
If you’ll make certain that this circuit is not up to what you need, then yes. But do the easy (and cheap) steps first.

Regards
George

*PS. Also check for proper functioning of the LED
__________________
["Second Law is a bitch." - SY] ["The Road To Heaven:Specify the performance & accept the design. The Road To Hell:Specify the design & accept the performance"-Bruno Putzeys]

Last edited by gpapag; 19th May 2011 at 01:17 PM.
  Reply With Quote
Old 19th May 2011, 07:22 PM   #26
gpapag is offline gpapag  Greece
diyAudio Member
 
gpapag's Avatar
 
Join Date: Nov 2002
Location: Athens-Greece
Regarding the main topic of this thread (LM317) PSU:

Have you seen this this?
(written by Eric Juaneda, poster on diyau.com)

High End Audio - LM317
__________________
["Second Law is a bitch." - SY] ["The Road To Heaven:Specify the performance & accept the design. The Road To Hell:Specify the design & accept the performance"-Bruno Putzeys]

Last edited by gpapag; 19th May 2011 at 07:25 PM.
  Reply With Quote
Old 19th May 2011, 09:39 PM   #27
gpapag is offline gpapag  Greece
diyAudio Member
 
gpapag's Avatar
 
Join Date: Nov 2002
Location: Athens-Greece
Tony
Now for the oscilloscope “Y out”

Some manufacturers -TEK is one of them- implement “Y out” as a buffered out signal of Input 2 channel. They pick the signal just after “Input 2 preamplifier”.
Lets call this “Case 1”

Others, implement “Y out” as a buffered out of whatever is channeled to the screen.
That can be Input 1, Input 2, Inverted Input 2, Input 1 minus Input 2, Input 1 plus Input 2 ect. They pick the signal just prior to the “Y main Amplifier”.
Lets call this “Case 2”

In all the oscilloscope schematics I’ve seen, at the pick-up point (both cases), the circuit is differential. So they use some form of “differential in to single ended out” circuit to bring out a single ended “Y out” which is specified as 50Ohm out.
The signal amplitude is 50mV per screen’s vertical division (mV/div) when the attached load is 50 Ohm, or 100mV/div open circuit.

Note here, that the maximum signal coming out the “Y out” can be much higher than simply (100mV/div) X 8 divisions. This is the case of high “Y amplification”, when only part of the waveform is visible on the screen, while the upper and the lower parts –or one of them only-are above and below the screen respectively, thus invisible.


At the schematics you provided for your oscilloscope:
For “Case 1”, I would suggest pins 7 and 8 of U152 (page 48/54 of pdf) as the pick-up points for the “differential to single ended converter” circuit which will bring out the single ended “Y out” See first attachment.

For “Case 2”, I would suggest the drive signal of the Y Output Amplifier (page 49/54 of pdf) as the pick-up points for the “differential to single ended converter” circuit. See second attachment.

Any objection or suggestion from members who have an opinion is welcome.

Tomorrow, I hope to discuss a little about the “differential to single ended converter”

Regards
George

EDIT: Actually, TEK provides a buffered Input 1 signal to "Y Out" and not Input 2 as I wrote above. My bad, sorry
Attached Images
File Type: jpg Case 1.JPG (140.7 KB, 628 views)
File Type: jpg Case 2.JPG (131.9 KB, 592 views)
__________________
["Second Law is a bitch." - SY] ["The Road To Heaven:Specify the performance & accept the design. The Road To Hell:Specify the design & accept the performance"-Bruno Putzeys]

Last edited by gpapag; 19th May 2011 at 09:49 PM.
  Reply With Quote
Old 20th May 2011, 07:51 AM   #28
just another
diyAudio Moderator
 
wintermute's Avatar
 
Join Date: Aug 2003
Location: Sydney
Blog Entries: 22
Thanks for all this George!! I searched for all sorts of variations on the 5514 but nothing turned up, a google search limiting results to Australia turns up nothing as well, must be one of those odd things where a part is just not stocked in a particular region.

I haven't bought a tl074 yet, I'll sim it in ltspice to check what happens first I didn't look far enough into the datasheet to see the example with unity gain, or the fact that it generally uses a +- supply (vendor site says up to 36V or +- 18V and I assumed that meant single rail was ok.

I've seen that LM317 site before. I thought his choice of tests was a bit odd. I don't understand why he would put 22ohms in series with the output cap, (I'm using .33ohms)

I think it interestng that he thought that the best sound was with the noisiest configuration, I wonder whether the noise is "adding" what he is percieving as the qualities mentioned... I doubt that my system (or my ears) have enough resolution to detect such things I'll be happy provided I get very low hum and hiss which is my goal, these are things I do hear and that annoy me.

I should post the results showing no resistance in series with the output cap, 0r33 and 0r66 I probably should also try a panasonic FC without the resistor (as I have a 1000uF one in my junk box). Can't post results now as on the way home on the train, but maybe later tonight

Tony.
__________________
Any intelligence I may appear to have is purely artificial!
Some of my photos

Last edited by wintermute; 20th May 2011 at 07:53 AM.
  Reply With Quote
Old 20th May 2011, 10:07 AM   #29
Piersma is offline Piersma  Netherlands
diyAudio Member
 
Piersma's Avatar
 
Join Date: Jan 2003
Location: Netherlands
Default Improved regulation

Dual LM317 configuration
Attached Images
File Type: png dual_LM317.PNG (36.3 KB, 578 views)
  Reply With Quote
Old 20th May 2011, 01:03 PM   #30
just another
diyAudio Moderator
 
wintermute's Avatar
 
Join Date: Aug 2003
Location: Sydney
Blog Entries: 22
Interesting configuration Piersma! I think I'll have to sim that because I don't understand how it works.... What's the value for R? 1 ohm or less??

As promised here are a few more measurements.

1st pic is comparison between 100uF cap on the output with 0r33 in series (left channel) and no resistor in series (right channel). Note that for some reason subsequent measurements with this cap did not get the flat noise floor and instead started to drop at higher frequencies.

2nd pic shows 100uF output cap with 0r33 left and 0r66 right channel in series.

3rd pic is one I did last night, it is a comparison of a single 1000uF cap as post rectifier filter (left channel) vs a single 10,000uF cap post rectifier filter.

4th pic is another one I did last night for another thread, a comparison between 4.7uF adj cap and 1000uF adj cap, with only a single 1000uF cap as the filter cap. whilst there is a difference it is marginal, and I think that the CRC is a better alternative.

Tony.
__________________
Any intelligence I may appear to have is purely artificial!
Some of my photos
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
Screen drive and other P-P experiments tubelab.com Tubes / Valves 23 6th July 2010 06:08 PM
Initial Experiments for Fun stuphi Multi-Way 17 4th June 2009 05:19 PM
My first horn experiments till Multi-Way 110 31st May 2004 02:55 PM
x-over experiments, diff. orders etc. SimontY Multi-Way 25 19th February 2004 11:19 PM


New To Site? Need Help?

All times are GMT. The time now is 11:54 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2